:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/49231451/GettyImages-518787306.0.jpg)
Wisconsin, which holds its presidential primary elections Tuesday, was one of the first states to adopt the direct primary at the statewide level back in 1904. The state adopted it in an effort to undermine the major political parties of the day. Yet it ended up not quite having that effect.
I explore this and some other examples of anti-party reforms in my new book The Inevitable Party: Why Attempts to Kill the Party System Fail and How They Weaken Democracy. Wisconsin's 1904 reform was actually one of the more dramatic anti-party moves. It marked the decline of a long tradition of party nominations made almost exclusively by party conventions under boss control. Most states would quickly follow Wisconsin's example, putting voters instead of party hacks and bosses in charge of nominating candidates.
The intent behind this reform was to break apart the state's intricate party machinery, which maintained a stranglehold on state politics. Robert La Follette, a Madison lawyer who would get elected governor in 1900 and later have a distinguished career in the US Senate, became one of the most vocal champions of the direct primary, seeing it as a way to break boss control. "No longer," the Progressive promised, "will there stand between the voter and the [elected] official a political machine with a complicated system of caucuses and conventions, by the easy manipulation of which it thwarts the will of the voter."
Complaints about the party system at the beginning of the 20th century were not entirely the same as today's, but still pretty similar. Reformers then were chiefly concerned about a party system "owing no allegiance and acknowledging no responsibility to the people," one that exerted too much control over politicians. They sought an election system that freed up leaders to be more independent. Through a primary, they reasoned, elected officials would be dependent upon the support of tens of thousands of ordinary voters rather than a few hundred party insiders. They'd be more inclined to follow the needs of their constituents than the wishes of the party bosses.
Interestingly, detractors of the primary reform sounded the same predictions. They were worried that the reform was "radical to a Populistic degree" and would destroy all party organization in Wisconsin. Shortly after the primary's adoption, opponents complained that "the names 'Republican' and 'Democrat' have no real meaning in Wisconsin today. ... Each member of Congress is a party by himself."
So what actually happened once the direct primary went into effect? The results are rather interesting. With the help of Keith Poole, I calculated ideal points for Wisconsin statehouse members from 1895 to 1911 based on their roll call voting patterns. The graph below shows those ideal points broken down by party, both before and after the adoption of the direct primary.
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/6283395/WI_polarization.0.png)
The graph suggests that Democrats (in the minority throughout this time period) were trending in a more moderate direction but then moved toward their extremes after the implementation of the direct primary. Majority Republicans, meanwhile, barely budged in their voting behavior. They stayed, on average, pretty consistent ideologically.
Further analysis shows that both parties became more internally cohesive; they voted more as unified party blocs under the direct primary. What's more, there's no evidence that legislators became better representatives of their districts under the direct primary. If anything, they became better representatives of their parties and worse district delegates.
How did this happen? Why did the reform designed to destroy party organizations only end up strengthening partisanship? To some extent, this appears to be yet another example of political reformers misunderstanding the nature of political parties. Parties can't simply be legislated away. They're run by very creative and adaptive people, and controlling the outcome of a primary election isn't necessarily that much harder than controlling the outcome of a party convention or caucus.
It's also worth acknowledging that many party reforms don't really end up throwing bosses out on the streets so much as empowering a different type of boss to take over. La Follette, in particular, was a powerful orator and highly skilled in new campaigning techniques, while earlier party leaders were better at manipulating convention machinery. La Follette was in many ways a strong party boss, just a different kind than had been in power previously with a different skill set. He engineered a change in election laws that advantaged leaders like him.
The nomination system we have today is a legacy of the work done by Progressive reformers in Wisconsin and other pioneering states at the turn of the 20th century. We have definitely made it a more democratic process. But putting the people in charge of nominating candidates hasn't necessarily resulted in candidates they're happier with, and it certainly hasn't killed off the parties.
